© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 7-RES-JB
    11. 23939
    Prev
    Next
    INDEX
    Explanations

    mentions of excessive consumption or indulgence in activities such as eating, drinking, media-watching, and spending

    oai_token-act-pair · gpt-3.5-turbo

    references to binge behavior, especially related to entertainment and consumption

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rossi
    -0.72
    ento
    -0.71
    ansas
    -0.68
    nces
    -0.68
     bureau
    -0.67
    orea
    -0.66
    rious
    -0.66
    netic
    -0.65
    çĭ
    -0.64
     Manila
    -0.64
    POSITIVE LOGITS
    vironment
    0.93
    uth
    0.92
    ervative
    0.87
    steen
    0.84
    terday
    0.81
    zyme
    0.77
    ð
    0.75
    Ò
    0.73
    imated
    0.73
    verend
    0.71
    Activations Density 0.086%

    No Known Activations